From issues-return-227377-archive-asf-public=cust-asf.ponee.io@spark.apache.org Tue May 21 07:34:48 2019 Return-Path: X-Original-To: archive-asf-public@cust-asf.ponee.io Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mail.apache.org (hermes.apache.org [207.244.88.153]) by mx-eu-01.ponee.io (Postfix) with SMTP id B598C18061A for ; Tue, 21 May 2019 09:34:47 +0200 (CEST) Received: (qmail 31324 invoked by uid 500); 21 May 2019 07:15:02 -0000 Mailing-List: contact issues-help@spark.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list issues@spark.apache.org Received: (qmail 31256 invoked by uid 99); 21 May 2019 07:15:01 -0000 Received: from mailrelay1-us-west.apache.org (HELO mailrelay1-us-west.apache.org) (209.188.14.139) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 21 May 2019 07:15:01 +0000 Received: from jira-lw-us.apache.org (unknown [207.244.88.139]) by mailrelay1-us-west.apache.org (ASF Mail Server at mailrelay1-us-west.apache.org) with ESMTP id A74DAE0CD9 for ; Tue, 21 May 2019 07:15:00 +0000 (UTC) Received: from jira-lw-us.apache.org (localhost [127.0.0.1]) by jira-lw-us.apache.org (ASF Mail Server at jira-lw-us.apache.org) with ESMTP id 637C925816 for ; Tue, 21 May 2019 07:15:00 +0000 (UTC) Date: Tue, 21 May 2019 07:15:00 +0000 (UTC) From: "Ajith S (JIRA)" To: issues@spark.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Updated] (SPARK-23626) Spark DAGScheduler scheduling performance hindered on JobSubmitted Event MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/SPARK-23626?page=3Dcom.atlassi= an.jira.plugin.system.issuetabpanels:all-tabpanel ] Ajith S updated SPARK-23626: ---------------------------- Affects Version/s: 3.0.0 > Spark DAGScheduler scheduling performance hindered on JobSubmitted Event > ------------------------------------------------------------------------ > > Key: SPARK-23626 > URL: https://issues.apache.org/jira/browse/SPARK-23626 > Project: Spark > Issue Type: Bug > Components: Scheduler > Affects Versions: 2.2.1, 2.3.3, 3.0.0 > Reporter: Ajith S > Priority: Major > Labels: bulk-closed > > DAGScheduler becomes a bottleneck in cluster when multiple JobSubmitted e= vents has to be processed as DAGSchedulerEventProcessLoop is single threade= d and it will block other tasks in queue like TaskCompletion. > The JobSubmitted event is time consuming depending on the nature of the j= ob (Example: calculating parent stage dependencies, shuffle dependencies, p= artitions) and thus it blocks all the events to be processed. > =C2=A0 > I see multiple JIRA referring to this behavior > https://issues.apache.org/jira/browse/SPARK-2647 > https://issues.apache.org/jira/browse/SPARK-4961 > =C2=A0 > Similarly in my cluster some jobs partition calculation is time consuming= (Similar to stack at SPARK-2647) hence it slows down the spark DAGSchedule= rEventProcessLoop which results in user jobs to slowdown, even if its tasks= are finished within seconds, as TaskCompletion Events are processed at a s= lower rate due to blockage. -- This message was sent by Atlassian JIRA (v7.6.3#76005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org For additional commands, e-mail: issues-help@spark.apache.org